At last, people admitting that there's an actual reason people don't want to switch to your thing when you make it unnecessarily incompatible with the old thing and don't solve any important problems.

http://alexgaynor.net/2013/dec/30/about-python-3/

Next stop: IPv6.

(Okay, only sorta kidding with that analogy.  I actually kind of thought python 3 might catch on eventually... :))